The Literature Machine revisited: spam lit, Calvino, OULIPO and conceptual literature

I posted in August last year about Spam Poetry, and used some examples from my own WordPress spam repository as the basis for some ‘found poetry’. In my previous post I offered another example of my own, ‘Update: an excavated fragment’, based on the dialogue between a computer user and the machine’s interface.

I have become aware, since that post last year, that the phenomenon of ‘spam poetry’ is quite well attested – I didn’t invent it after all, though at the time I misguidedly thought I might have done. The rest of this post will provide more context.


In July of this year I posted about VOLVELLES: mechanical devices that might be called early ‘paper computers’, primitive expert systems or thinking machines, usually employed for calculating or generating information or texts. These can be traced back to the ancient east, but in the west to Ramon Lull (the Ars combinatoria), and later, Leibniz, Kircher, and so on.

Swift in Gulliver’s Travels may have been satirising Leibniz in his ‘permutational machine’ at the Academy of Lagado. Centos, bibliomancy and later literary techniques like cutup and permutational poetry can also trace their origins to such ‘literary machines’.

This is the title of the book of essays by Italo Calvino, about which I also posted in July this year, with a focus on ‘Cybernetics and Ghosts’. In this essay Calvino considers the difference between the ‘random text generator’ and what might be called the Literary Machine: the procedure which bypasses individual human inspiration by using ‘combinatorial play’ to generate texts through the permutation of a restricted number of elements and functions. French avant-garde groups like OULIPO have experimented with such methods for decades now, harking back to the Surrealists with their use of self-imposed constraints in the production of literary texts.

Which brings me back to my own ‘spam poems’ cited earlier. A little judicious searching online will readily take you to more detailed information on the following (Wikipedia, for example, has much more on this, with examples and links):

SPAM POETRY: here the involvement of an author in the production of literature, as Calvino speculated, has become discretionary. It works on similar principles to automatic writing (Yeats was an aficionado of this, with its mystical/supernatural overtones), which was also favoured by Freud and the Surrealists as a means of tapping spontaneously into the unconscious to produce ‘psychography’. Aleatoric writing is a related concept: writing produced on the principle of accidental or chance language choices.

A key concept in such text generation is what OULIPO called ‘clinamen’ (and a near-translation, ‘swerve’): an arcane term for the classical notion of ‘primordial anti-constraint’. Creation (of a text) is rendered possible, in an ordered, logical, rational universe, by the introduction of chance. Harry Mathews’ algorithm applied to Queneau’s cutup sonnet sequence would be an example of ‘coercing’ the potentiality of texts into existence. Language is exploited through the use of matrices, and computers make this process millions times quicker and more productive than old-style cut-and-paste.

Spam is usually created by computer programs which randomly copy extracts from internet texts (Burroughs and his predecessor, Brion Gysin, called this ‘paratext’), reassemble them and use them to try to smuggle marketing or other unsolicited messages through the filters of blogs and other websites. They try to trick the spam filters into thinking that the ‘text’ thus generated has been created by human hands, for the filters usually lack the sophistication to distinguish gibberish from texts which have semantic coherence.

In brief, I’d suggest, if you’re interested in pursuing this ‘mechanical literature’, you research similar ‘genres’:

FLARF: text produced by the anonymised and reshuffled errancies of various machine protocols (Wikipedia)

SPAM LIT: the site called UbuWeb (‘an anthology of conceptual writing’) has a wealth of useful examples, articles and links.

See also:

SPOETRY, WORD SALAD, GOOGLISM, INFORMATIONIST POETRY, THE L-A-N-G-U-A-G-E POETS (most in various ways use the detritus of the internet as a source for material for recombining or regenerating texts).

I’d have to conclude, however, that the literary results are decidedly uneven. There are occasional felicitous juxtapositions created through the use of these techniques (and I’d like to think there were some among the examples I produced myself, in which I intervened and selected from the morass of verbiage to create something rather more…orderly and, I hope, interesting). But much of it is doggerel.

Italo Calvino and the Literature Machine

Italo Calvino: 1923-1985

In my last post I mentioned taking down from my much-reduced bookshelf space (builders at work) Italo Calvino’s collection of essays The Literature Machine (Vintage, 1997), translated from the Italian by Patrick Creagh. In the first Part he writes about literature in general, especially in relation to narratology, structural(ist) linguistics/semantics/semiology, the ‘intellectual scorn’ of the Collège de Pataphysique and its offshoot of mathematical/mechanic literati the Oulipo, founded by Raymond Queneau – an avant garde group which he was invited to join when he was in Paris in 1968. His own work had veered dramatically away from the realist tradition of fiction to which he adhered in his earlier works. I hope to write in more detail about Calvino’s literary achievement in a later post.

Calvino interviewed by L. Salori for RAI, 1958

Calvino interviewed by L. Salori for RAI, 1958

He also writes here on literature and philosophy, as a projection of desire, on comedy, eroticism and fantasy, on the putative audience for whom authors write, on ‘levels of reality’ and on politics (having fought against the fascists in Italy during WWII and joined the CP, he left it in 1957 in a reaction against the Soviet invasion of Hungary the previous year, and the revelations of Stalin’s excesses).

Part Two opens with the famous essay ‘Why Read the Classics?’, in which he defines a classic as ‘a book that has never finished saying what it has to say’, which is rather more pertinent than Mark Twain’s (admittedly funnier )‘a book that everyone wants to have read, but no-one wants to read’. He ends with this fine quotation from the Romanian philosopher Emil Cioran (1911-95) to illustrate his point that classics shouldn’t serve any particular ‘purpose’:

While they were preparing the hemlock, Socrates was learning a tune on the flute. ‘What good will it do you,’ they asked, ‘to know this tune before you die?’


(Interesting that ex-communist Calvino should cite this former admirer of Hitler – although Cioran did after the war recant his pro-fascist views). The rest of Part Two deals mostly with individual authors and texts: The Odyssey, Ovid, Candide, Balzac and the city, Stendhal, etc.


Labyrinth at Tintagel, Cornwall

Labyrinth at Tintagel, Cornwall

But I should like to return to the opening essay, ‘Cybernetics and Ghosts’. After discussing the evolutionary development of literary critical theory, he considers the ways in which groups such as the anarcho-literary Oulipo artists (mentioned above) utilise mathematical patterns and self-imposed constraints when generating literary texts. I intend posting about this in more detail in a later piece.

The next logical step, he suggests, is to create computer programs that are capable of literary production. Not random text generators (there are plenty of these around online now) which work on the principle of ‘destructuralization of form’; what he posits is a need for ‘the production of disorder’, a program that reacts against its ‘preceding production of order’. Such programs would reject traditionalism and ‘propose new ways of writing’ – the Romantic notion of ‘inspiration’, the individual poet’s Coleridgean ‘afflatus’, the intuitive impulse arising from … where? The unconscious? Who knows. We need for the figure of the author (‘that spoilt child of ignorance’, Calvino calls him) to disappear, and to privilege that of the reader, who will know ‘that the author is a machine, and will know how this machine works’.

If his reader, he continues, is suspicious of his motives in consigning the inspirational poet-author to the bin, he’d respond that all writing is ‘simply a process of combination among given elements’. If it’s ‘merely the permutation of a restricted number of elements and functions’ isn’t this too restrictive? Literature constantly tries ‘to say something it cannot say, something it does not know, and that no one could ever know’. Such statements have never been arranged with the words in that particular order before. So the struggle of literature is ‘a struggle to escape from the confines of language’- the ‘call and attraction of what is not in the dictionary’.

Lit Machine coverThis takes us into the territory of myth and ritual, and taboos, the ‘ban on mentioning something’ – certain names or prohibitions. Literature constantly pushes against these boundaries of what can’t be said, ‘to an invention that is a reinvention of words and stories that have been banished from the individual or collective memory’.

Literature redeems these forbidden territories, giving voice to what’s remained unexpressed in ‘the social or individual unconscious…The more enlightened our houses are, the more their walls ooze ghosts’. The Surrealists found a means of using word and image association to find an opposing rationale to intellectual logic.

This leads Calvino to an exploration of Ernst Kris’s development of Freud’s study of word-play as a ‘possible aesthetics of psychoanalysis’. Puns and  jokes derive from playing with permutations and transformations in language, from ‘combinatorial play’. By way of such combinatorial mechanisms a new combination is arrived at with ‘an unexpected meaning or unforeseen effect which the conscious mind would not have arrived at deliberately’. This is ‘an unconscious meaning, or at least the premonition of an unconscious meaning’. This shock of novel meaning ‘occurs only if the writing machine is surrounded by the hidden ghosts of the individual and society’; the outcome should be ‘a terrible revelation: a myth’, to be recited in secret, and stumbled upon only by ‘playing around with narrative functions’.

Borges (left) with fellow Argentine writer Ernesto Sabato in 1975

Borges (left) with fellow Argentine writer Ernesto Sabato in 1975

From here it’s a short step to the labyrinthine narratives that Borges inherited from ancient sources, such as the Chinese-box narrative-within-a-narrative. Such literature does not enlighten the reader: we are disorientated by it, our world is defamiliarised. By experiencing bewilderment we are able to liberate ourselves from it. A labyrinth is designed to be entered into with a view to becoming lost. By reconstructing its plan we destroy it. Literature thereby becomes a means of demonstrating that ‘the world is essentially impenetrable, that any communication is impossible.’ The labyrinth becomes ‘a facsimile of the world and of society’ (Enzenberger).

The combinatorial mathematical game acts as ‘a challenge to understand the world or as a dissuasion from understanding it’. The reader’s role is therefore decisive: ‘It is up to the reader to see to it that literature exerts its critical force, and this can occur independently of the author’s intentions’. Elsewhere, I suppose, this is what is known as the Intentionalist Fallacy.

There is a manic elegance in this surreal or labyrinthine logic. It’s found in so much of the fiction of the later 20C, from Kafka to Beckett, Borges to the stories of Barthelme and Coover, and ultimately in Pynchon and, perhaps, DeLillo.

In a later post I shall examine more closely this notion of mechanical constraints as a mechanism for aiding (by constraining) the author in generating texts; after all, what could be more artificially limiting than composing a traditional Shakespearean sonnet, with its strict metre, rhyme scheme and stanzaic pattern? On the other hand, is omitting from your novel all words containing the letter ‘e’ a particularly sound idea? All literary texts involve some element of willed omission and selection…

Endnote: The New York Times started a column in 2011 called ‘The Mechanic Muse’, which focuses on issues related to those raised by Calvino and Oulipo; the inaugural essay in the series by Kathryn Schulz (24.06.11) profiled the ‘distant reading’ project (a deliberate reversal of Leavisite ‘close reading’ discussed in my previous post) of Franco Moretti at his Stanford Literary Lab. Link here.)

(Images in the public domain via WikiCommons)